Assessment of RELAP5/MOD3.2.2 Gamma against ABB Atom 3×3-Rod Bundle Reflooding Tests

The results of the ABB Atom 3×3-Rod Bundle Reflooding Tests were used for assessment of the reflooding model used in RELAP5/MOD3.2.2 Gamma version. The assessment calculations were performed using the default calculation model options implemented in the code.The tests were performed to investigate the effects of different spacer grid designs on heat transfer during the reflooding period of a pressurized water reactor loss-of-coolant accident (LOCA). The tests were conducted under low-pressure and low-flow (LPLF) conditions using a PWR-type 3×3-rod bundle with full-length indirectly electrically heated, stepped cosine axial power-shaped heater rods. Three different spacer grid configurations were studied: spacer grids without mixing vanes, mixing vane spacer grids, and mixing vane spacer grids together with intermediate flow mixers (IFM).A total of 36 tests with different spacer grid configurations were calculated. For two selected basic tests with non-mixing spacer grids an extended comparison of calculated and measured parameters is presented. The comparison of the predicted and measured maximal cladding temperatures and quench times, which are the most important parameters in licensing calculations, is presented for all the performed tests.The assessment calculations were preceded by nodalization, time step, and moving mesh studies.The RELAP5/MOD3.2.2 Gamma code was found to still have several deficiencies in the reflood model. The calculation results show a satisfactory agreement with experimental inner peak cladding temperature, however the predicted temperature turn-around times and quench times are significantly too short. The results also show a significant over-prediction of the reflood heat transfer and the vapour temperatures. The void profile downstream the quench front is not correctly predicted either. Finally, the present reflood model does not properly reflect the effects of spacer grids on the reflood heat transfer.In spite of these deficiencies the improvements incorporated into RELAP5/MOD3.2 by the Paul Scherrer Institute (PSI) eliminated the unphysical behaviors such as continuous cooling without clear turn-around temperature and no visible quenching phenomena, which were shown in the reflood calculations by means of the RELAP5/MOD3.1 code.